Ask Claude to Explain and Document Cryptic Regex Patterns

Regex is easy to write once and impossible to understand six months later. When you inherit a pattern you can't decipher, paste it to Claude and ask for a plain-English breakdown plus a commented version using the verbose flag — then you'll actually understand it when you need to change it.

Ask Claude:
"Explain this regex and rewrite it with inline comments using the verbose/x flag:
^(?:[a-z0-9!#$%&'*+/=?^_`{|}~-]+(?:\.[a-z0-9!#$%&'*+/=?^_`{|}~-]+)*|\"...\")@..."

Claude produces a readable, commented version:

import re

email_pattern = re.compile(r"""
    ^                              # Start of string
    (?:
        [a-z0-9!#$%&'*+/=?^_`{|}~-]+      # Local part: allowed characters
        (?:\.[a-z0-9!#$%&'*+/=?^_`{|}~-]+)* # Dots allowed, not consecutively
    )
    @                              # Literal @ separator
    (?:[a-z0-9](?:[a-z0-9-]*[a-z0-9])?\.)+  # Domain labels with dots
    [a-z0-9](?:[a-z0-9-]*[a-z0-9])?$        # Top-level domain
""", re.VERBOSE | re.IGNORECASE)

Claude will also point out edge cases the regex doesn't handle — in this case, it doesn't match IP address literals or some valid RFC 5322 formats — and suggest whether a simpler pattern would cover your actual use case just as well.

This pairs well with the tip on generating new regex from descriptions: use Claude to generate, and use Claude to document.

If you can't explain a regex, Claude can — and it'll tell you how to make it better.